Search

Using coding in phase sensitive optical time domain reflectometry (ϕ-OTDR) system has long been pursued. However, since no single sequence with ideal aperiodic autocorrelation has been found, sidelobes of aperiodic autocorrelation induce crosstalk among sensors. Finding coding schemes with as low crosstalk as possible is very meaningful. In this paper, a novel ϕ-OTDR scheme with coding using polyphase unimodular sequence is proposed. Theoretical analysis of crosstalk due to coding sequence is presented at first. It shows that crosstalk is upper-bounded by sidelobes of aperiodic autocorrelation which is also an effective indicator of crosstalk level. Polyphase unimodular sequence with very low aperiodic autocorrelation sidelobes has then been searched through optimization of the Frank sequence using gradient descent algorithm. The merit factor (MF) of the obtained sequence is more than 2 times of that of the original Frank sequence. Compared to the pseudorandom binary sequence (PRBS), the MF of the obtained sequence is more than 10 dB higher. The evolution of MF of the obtained sequence at various stages and different transmission distance in a ϕ-OTDR system is demonstrated, which confirms that the sequence can support long-range sensing even though phases of the sequence are densely spaced. A ϕ-OTDR system with a weak fiber Bragg grating (WFBG) array is then set up to quantify crosstalk due to the polyphase unimodular sequence. In comparison to PRBS, experiment results confirm that the polyphase unimodular sequence has much lower crosstalk level and enhance the crosstalk rejection ratio by more than 10 dB.

Interference fading problem is an inevitable challenge in distributed acoustic sensing (DAS) system, and the uncertainty of the dynamic external environment further increases the difficulty of denoising. Based on the conventional single-domain fading suppression methods, we propose the concept of integrated fading suppression (IFS) algorithm set, which provides four denoising approaches and each of them is composed of at least two independent techniques. The combination of multiple single-domain denoising methods enhances the reliability of the proposed method. In an experiment involving 30 sets of data, the IFS algorithm set succeeded in completely eliminating all fading noise. In addition, the demodulation results show that compared to only applying a bandpass filter, these four IFS algorithms could improve the intensity signal-to-noise ratio by an average of 5.82 dB, while decreasing the amplitude fluctuation range by 31.84 dB on average. Benefiting from its powerful fading suppression capability, the average strain noise of the demodulated results obtained by the proposed algorithm set goes down to 0.47 nϵ. It is worth noting that the IFS algorithm set with four denoising methods is highly flexible and could meet different requirements and preferences. With the proposed IFS algorithm set, it is expected that the DAS system could be thoroughly free from the influence of interference fading.